Carbon Dioxide Capture with Amine Functionlized Graphene Oxide

University of Mississippi eGrove

Electronic Theses and Dissertations Graduate School

2015

Carbon Dioxide Capture With Amine Functionlized Graphene Oxide

Renee Zeleszki University of Mississippi

Follow this and additional works at: https://egrove.olemiss.edu/etd

Part of the Chemical Engineering Commons

Recommended Citation Zeleszki, Renee, "Carbon Dioxide Capture With Amine Functionlized Graphene Oxide" (2015). Electronic Theses and Dissertations. 917. https://egrove.olemiss.edu/etd/917

This Thesis is brought to you for free and open access by the Graduate School at eGrove. It has been accepted for inclusion in Electronic Theses and Dissertations by an authorized administrator of eGrove. For more information, please contact [email protected]. CARBON DIOXIDE CAPTURE WITH AMINE FUNCTIONLIZED GRAPHENE OXIDE

A Thesis Presented in partial fulfilment of requirements For the degree of Master of Science In the Department of Chemical Engineering The University of Mississippi

RENEE RAN WEI ZELESZKI

August 2015

Amine functionalized graphene oxide was used as sorbent in order to achieve carbon dioxide capture. In this study, micro size graphite flakes was prepared according to Modified

Hummers method and “Hummers’ method with additional KMnO4”. Graphene oxide was then functionalized with ethylenediamine (EDA). Different purification methods for graphene oxide were studied as well as different functionalization method times. Graphite oxide and functionalized graphene oxide has been analyzed with Fourier transform infrared spectroscopy

(FTIR) for chemical structure. Thermal gravimetric analysis (TGA) was used to test CO2 capture capacity. Functionalized graphene oxide that has been purified with 30 washes via centrifuge with 72 hour amine functionalization method time had greatest CO2 capture potential with 1.38% weight increase.

ACKNOWLEDGMENTS

This thesis is dedicated to Dr. Chen and CO2 capture group. I want to thank everyone who has helped me with this paper. In particular, I thank Dr. Chen and Dr. Mattern for the idea and help.

iii

TABLE OF CONTENTS

ABSTRACT ...... ii

ACKNOWLEDGEMENTS ...... iii

LIST OF TABLES ...... v

LIST OF FIGURES ...... vi

INTRODUCTION ...... 1

EXPERIMENTAL PROCEDURE ...... 9

RESULTS AND DISCUSSION ...... 15

CONCLUSION AND FUTURE WORK ...... 27

REFERENCES ...... 31

APPENDIX ...... 34

VITA ...... 48

LIST OF TABLES

TABLE 1 pH of graphite oxide after each wash ...... 36

TABLE 2 Weight gain with correction ...... 42

LIST OF FIGURES FIGURE 1...... 2

FIGURE 2...... 3

FIGURE 3...... 5

FIGURE 4 ...... 7

FIGURE 5...... 8

FIGURE 6 ...... 10

FIGURE 7 ...... 11

FIGURE 8 ...... 12

FIGURE 9...... 13

FIGURE 10 ...... 15

FIGURE 11 ...... 16

FIGURE 12 ...... 17

FIGURE 13 ...... 19

FIGURE 14 ...... 20

FIGURE 15 ...... 21

FIGURE 16 ...... 22

FIGURE 17 ...... 23

FIGURE 18 ...... 24

FIGURE 19 ...... 25

FIGURE 20 ...... 26

FIGURE 21 ...... 28

FIGURE 22 ...... 36

FIGURE 23 ...... 38

FIGURE 24 ...... 40

FIGURE 25 ...... 41

FIGURE 26 ...... 42

FIGURE 27 ...... 44

FIGURE 28 ...... 46

FIGURE 29 ...... 47

CHAPTER I

INTRODUCTION

I.1. Carbon Dioxide and its effect

Carbon Dioxide is one of the most well-known greenhouse gases. Excess CO2 has been emitted into the atmosphere due to human activities. CO2 is currently around 400 ppm (0.04%) in air compared to 1950 which was at around 283 ppm (0.03%) . Before 1950, within the past

650,000 years, CO2 has never exceeded 300ppm. There is strong evidence support of global warming such as increasing sea level, melting of polar ice cap in the North Pole and South Pole, and average global temperature has been rising with more extreme high temperature and less snow (1). Therefore, it is important to reduce CO2 emission, and thus slow down the greenhouse effect.

Figure 1 From nasa.gov “This graph, based on the comparison of atmospheric samples contained in ice cores and more recent direct measurements, provides evidence that atmospheric CO2 has increased since the Industrial Revolution. (Credit: Vostok ice core data/J.R. Petit et al.”(1) As shown in Figure 2, human activities such as burning fossil fuel make up 94% of total

CO2 emission—where electricity and industry make up 52% of the total emission. The goal of the current work is to target the industry CO2 emission and capture then store/utilize captured

CO2(2).

Figure 2 From epa.gov “All emission estimates from the Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2012.”（2）

I.2. Some current solutions:

I.2.1. Membrane separation:

Membrane such as polymeric and metallic membranes can be used to separate CO2 from

H2. Polymeric membranes are cheap, but there could be an issue when used with flue gas due to high temperature and potential corrosion from harsh chemicals. Metallic membranes fall into one of the two categories: dense or porous. A dense membrane contains a thin layer of metal and offers high selectivity for hydrogen or oxygen depending on the metal. However, unlike a porous membrane, dense membranes have low flux and in turn are not feasible for large scale gas separation. A porous membrane has a porous thin top layer and a porous thicker support layer.

Porous metallic membrane such as a silica membrane can stand harsher environment such as high temperature and corrosive chemicals (3).

I.2.2. Adsorption:

Adsorption refers to the process of a chemical (gas/liquid) attaching itself to the surface of a sorbent. With highly porous material such as activated carbon (surface area as high as 500m2

2 /g-1000m /g), CO2 is trapped inside the pores. The adsorption rate depends on the pore size; a larger surface area might not result in better CO2 capture capacity since CO2 is relatively large compared to H2 and N2 (3).

I.2.3. Absorption:

Absorption refers to the process where a chemical enters the interior of a sorbent. This includes using aqueous ammonium and aminated materials to react with CO2 resulting in CO2 chemically bounded to the sorbent. This is related to the current work—using amine functionalized graphene oxide (GO) as a sorbent for CO2 capture (3).

I.3. Graphite Oxide

I.3.1. History of Graphite Oxide

Graphite oxide is a compound that is made out of carbon, oxygen, and hydrogen. It was first discovered by Benjamin C. Brodie in 1859 by reacting graphite with potassium chlorate and nitric acid. In 1957, Hummers and Offeman developed The Hummers’ Method which uses sulfuric acid, sodium nitrate, and potassium permanganate to oxidize graphite, and it is currently the most used method to date. There are many different studies and modifications to Hummers’

Method in order to improve oxidation rate and reduce toxic gas byproduct. The method that is used in this paper was developed by James M. Tour el al. from Rice University (4).

I.3.2. Structure of Graphite Oxide

Graphite oxide is made of stacked layers of graphite sheets. It contains three different functional groups: epoxy, hydroxyl, and carboxyl groups; epoxy and hydroxyl are located on the basal plane, and carboxyl groups are located on the edges.

Figure 3 Lerf-Klinowski model of graphene oxide from literature (6)

I.3.3. Graphite Oxide to Graphene Oxide (denoted GO):

Graphite oxide is hydrophilic, which means it can be easily dispersed in water and other common polar solvents such as acetone, methanol, and ethanol. With the help of a physical force such as ultra sound or excessive stirring, graphite oxide can be broken down into graphene oxide.

Chemically, graphene oxide and graphite oxide are similar, but physically, they are very different.

Graphite oxide is neatly stacked and each layer is held together while graphene oxide is single to few layers thick resulting in greater dispersibility and larger surface area (13).

I.4. Current Work

The two functional groups that are used for CO2 capture are epoxy and carboxyl groups.

The carbon to oxygen ratio of graphene oxide is 2.1-2.9:1. The functional groups epoxy and carboxyl make up around 40% of total oxygen (6). Both groups are functionalized by ethylenediamine (EDA) with the help of leaving groups:1-ethyl-3-carbodiimide (EDC) and N- hydroxysuccinimide (NHS) (Figure 4).

Figure 4

Functionalized GO can be used for CO2 capture by reacting CO2 with aminated functional groups, as shown in Figure 5. If amines are attached to most of the epoxy and carboxyl groups, there should be an increased weight of 10%-15% from nitrogen. Amines have shown to have around 50 molar percent CO2 capture capacity, and thus graphene oxide should have a total of 5-7% weight increase.

Figure 5. Above: CO2 reacts with aminated epoxy group. Below: CO2 reacts with aminated carboxyl group

Together with ultrasound and water, the captured CO2 can be utilized to increase the heating value of biochar (created by burning biomass). The Kolbe-Schmitt reaction allows CO2 to react with biochar to increase its hydrogen and carbon content since H2O acts like a hydrogen donor and TiO2 is naturally present (helps aide photoreduction). Additionally, biochar exhibited increased internal surface area, significant removal of minerals and most importantly improved heating value of 50% (11).

CHAPTER II

EXPERIMENTAL PROCEDURE

II.1. Production of Graphite Oxide

II.1.1. Method 1, Modified Hummers’ Method

o In a 500ml flask, 23 ml of 90% sulfuric acid (H2SO4) was chilled to 0 C by placing the flask in ice water bath for 30 minutes. One gram of graphite flakes (Asbury carbons, Micro 450,

~2µm) was added to chilled H2SO4. Stirring the mixture in the ice bath, 3 g of potassium permanganate (KMnO4) was added slowly (~0.01g at time) to the mixture while the temperature was carefully monitored to keep the reaction under 20oC. The flask was then removed from the ice bath and stirred at 35oC for two hours. After 2 hours, 23 ml of water was added to the mixture slowly (1 drop at time) then stirred for 15 minutes at room temperature; afterwards 140 ml of water was added. Next, 10 ml of 10% hydrogen peroxide (H2O2) was added to terminate the reaction and graphite oxide was produced. Graphite oxide mixture was poured into four 50 ml centrifuge tubes and centrifuged at 5000 rpm for five minutes. The filtrate was carefully poured out, and washed with 40 ml of water per tube by centrifuge (liquid was added to the tubes, shaken, centrifuged for five minutes then the filtrate was carefully poured out – this process is denoted as 1 washing cycle in this paper). The process was repeated until the filtrate pH stabilized and graphite oxide was then dried under a fume hood overnight, and finally put in a vacuum oven until the weight stabilized, to create purified graphite oxide (4).

II.1.2 Method 2, Hummers’ method with additional KMnO4:

II.1.2.1.Oxidization:

o In a 500 ml flask, 23 ml of 90% H2SO4 was chilled to 0 C by placing the flask in ice water bath for 30 minutes while one gram of graphite flakes (Asbury carbons, Micro 450, ~2µm) was added to chilled H2SO4. Stirring the mixture in the ice bath, 0.5 g of sodium nitrate (NaNO3) was added to the mixture followed by slow addition of 3 g of KMnO4 while the temperature was carefully monitored to keep the reaction under 20oC. The flask was then removed from the water

o bath and stirred at 35 C for 7 hours. After 7 hours, an additional 3 g of KMnO4 was added to the mixture then stirred for 12 hours at 35oC (Figure 6).

Figure 6

Next, the mixture was poured over 130 g of ice, and 1 ml of H2O2 was added to terminate the reaction. Bubbles were observed and mixture turned from brown (coffee color) to caramel yellow (Figure 7).

Figure 7

II.1.2.2.1. Purification 1:

. Graphite oxide solution was poured into four 50 ml centrifuge tubes and centrifuged at

5000 rpm for five minutes. The filtrate was carefully poured out, and 40 ml (per tube) of water was added to the tubes after which were shaken (all washing cycles require 40 ml of additional liquid per tube). The washing process was repeated with of 30% HCl, 100% ethanol, and 100% ethanol (again) respectively. The mixture was washed with water 6 times; then dispersed in 200

11 ml water and stirred overnight. The washing procedure was repeated for the next day with 40 ml of water 10 times. On the third day, graphite oxide was washed with 40 ml of water eight times followed by washing with 40 ml of 100% ethanol twice for a total of 30 washes over three days.

The sample was then dried under a fume hood overnight, and finally put in a vacuum oven until the weight stabilized to create purified graphite oxide (4).

II.1.2.2.2. Purification 2 (Continued from Method 2 oxidation):

Graphite oxide solution (after oxidation) was poured into four 50ml centrifuge tubes and centrifuged at 5000 rpm for five minutes. The filtrate was carefully poured out, and washed by centrifuge with 30% HCl, 40 ml of 100% ethanol, and 100% ethanol (again) respectively for a total of four washes (40 ml liquid /tube during each wash). The sample was then dried under a fume hood overnight, and finally put in a vacuum oven until the weight stabilized to create purified graphite oxide (4).

Figure 8 Dried purified Graphite Oxide

II.2. Functionalization:

Purified graphite oxide (0.25 g ) was dispersed in 200 ml water, and then ultrasound was applied for 20 minutes to create graphene oxide (GO). N-Hydroxysuccinimide (NHS, 0.5 g) and

1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC, 0.5g) were added, and the mixture was stirred for 24 hours. The mixture was evenly disturbed in four centrifuge tubes then centrifuged at 5000 rpm for 5 minutes and washed by centrifuge with methanol twice. GO was then dispersed in methanol and 0.5 g of ethylenediamine (EDA) was added. Next, the mixture was stirred and one-third was removed after each time interval for three different lengths of time: 24 hours, 72 hours and 144 hours. Each product was evenly distributed to two 50 ml centrifuge tubes and centrifuged at 5000 rpm for 5 minutes and filtrate was then removed. The solid

(functionalized GO) from each tube was then washed by centrifuge with methanol, and then with acetone twice (40 ml liquid/tube). Finally the functionalized GO was dried overnight (8).

Figure 9 Dried Functionalized GO

II.3. Analysis

Thermogravimetric Analysis (TGA):

TA Q500 from TA Instruments was used. Two types of gas were used--carrier gas and

CO2. Both types of gas passed through a filter (to remove moisture and other impurities) and entered the balance pan at a flowrate of 40ml/min. For method 1, the carrier gas used was Grade

4.5 N2 (>99.995% pure), and for method 2, Grade 5 (> 99.999% pure) helium was used as the carrier gas. Gas #1 was set up to be the carrier gas, and gas #2 was set up as CO2 (high purity, >99.999% pure). Gas 1 was selected and increase (ramp program) at 20.0oC/min to

80.00oC followed by 20 minutes of dwell time at 80.00oC (isothermal program) or until the weight stabilized in order to remove impurities such as water and absorbed CO2. The sample was then cooled (ramp program) by 20.0oC/min to 60.00oC followed by the isothermal program for

20 minutes or until the weight stabilized. Next, gas #2 was selected and the process was under the isothermal program for additional 30 minutes or until the weight stabilized. The percent weight difference was calculated by using Universal Analysis.

CHAPTER III

RESULTS AND DISCUSSION

III.1. Method 1

The literature for method 1(7) stated that graphite oxide should reach a neutral pH; therefore, the pH of the filtrate was measured as shown in Figure 10. Although pH decreased overall, the efficiency of each washing decreased since pH never reached the desired value after

20 washes. Clearly, graphite oxide would not be able to reach a neutral pH within a reasonable number of washes.

Figure 10 pH of graphite oxide after each wash.

After washing and drying the sample, it was then functionalized with EDA and stirred for

72 hours. Graphite oxide was compared with functionalized GO via Fourier transform infrared spectroscopy (FTIR spectroscopy). The graphite oxide sample exhibited all three functional groups: OH (hydroxyl group) group at 3173cm-1, epoxy group at 1273 cm-1 & 1039 cm-1, and the

C=O (carboxyl group) at 1619 cm-1 (Figure 11). Therefore, graphite was successfully oxidized and all three functional groups were present. After functionalizing, we expected the disappearance of the epoxy group and the carboxyl group and an addition of peak at 3350 cm-1

(N-H stretch), but there was no significant difference between the two which might indicate nitrogen was not attached. Thus GO was not successfully functionalized with amine (9).

Figure 11 , FTIR spectrum of graphite oxide and Functionalized GO.

Functionalized GO was then analyzed with TGA. As shown in Figure 12, there was very little weight increase after the gas was switched. There was also a disturbance during the switch.

This might be due to the molecular weight of CO2 at 44 g/mol which is much heavier than the carrier gas N2 at 28 g/mol. Considering both gas were at the same volumetric flowrate (40 ml/min), the difference in buoyancy force can create a decrease in weight before the sorbent gets a chance to absorb CO2. Note there was an increase in weight during the nitrogen stage which could indicate that N2 was competing with CO2 and was absorbed instead; consequently, the absorption rate of CO2 was decreased. Therefore, in method 2, helium was used instead to prevent the N2/CO2 competition and a blank run was prepared prior to the sample runs in order to correct any weight difference due to buoyancy force.

Method 1, Functionalized GO 120 100.20%

100.00% 100 Temperature % Weight 99.80%

80 99.60%

C o 99.40% 60 <--Switched to CO2

99.20% Weight Weight %

Temperature, 40 99.00%

98.80% 20 98.60%

0 98.40% 0 10 20 30 40 50 60 Time, Minutes

Figure 12 TGA Result of Method 1 Functionalized GO

In order to counterbalance the buoyancy force, two blank samples were run and the percent weight loss was calculated (Figure 24 & 25). One sample was run with only the pan and the other sample was run with pan and an inert material (glass beads) that weighted around 20 mg. There was a 0.0337 mg weight loss in the empty pan run and a 0.0344 mg weight loss in the weighted sample run. The percent error between the two methods was 2% which can be considered insignificant and may be neglected. Therefore, 0.0337mg was added to the raw weight gain for method 2. Note this weight loss does not apply to method 1 due to the difference in carrier gas. The molecular weight of N2 (28 g/mol) is much higher than He (2 g/mol).

III.2. Method 2:

In method 2, two purification procedures were used. The first procedure was similar to method 1 where graphene oxide was purified by washing until the pH stabilized (sample was denoted 30W graphite oxide). In the second procedure, only 4 washes were used (including 30%

HCl); therefore, pH was not measured since pH was no longer a good indication for purity

(sample was denoted 4W graphite oxide). As shown in Figure 13, they have very similar peaks which could indicate the presence of all three functional groups. The broad strong peak from

3000 cm-1 —3500 cm-1 was OH group, sharp strong peaked at 1220 cm-1 and 1057 cm-1 signified epoxy group, and the sharp strong peak at 1620cm-1 represented the C=O (carboxyl group). The peaks at the far left could be noise and should be ignored (9). Note there was a peak intensity difference in the OH group between the two methods. In the 30W graphite oxide, the OH peak was more intensive compared to other groups which means more OH bonds existed in comparison. In the 4W graphite oxide, the intensity of each peak was about equal. This means there were more hydroxyl bounds in the 30W graphite oxide than the 4W graphite oxide—in

18 other words, there were less carboxylic acid and epoxy bounds in the 30W graphite oxide compared to the 4W graphite oxide.

Figure 13, FTIR spectrum of graphene oxide where green line represents 4W graphite oxide and red line represents 30W graphite oxide III.2.1. Method 2 with Purification 1 (30W GO)

Purified graphite oxide samples were functionalized separately. With purification method

1 (30W graphite oxide), three different method times were used: 24 hours, 72 hours, and 144 hours (denoted 30W GO F1, 30W GO F2, and 30W GO F3 respectively). All the methods were analyzed with FTIR spectroscopy. As shown in Figure 14, all three method times have resulted in identical peaks. For 30W GO F2 (72 hours) and 30 W GO F3 (144 hours), the lines overlapped each other.

Figure 14 FTIR spectrum of functionalized GO

All three samples were tested for CO2 capture capacity. As shown in Figure 15, 30W GO

F1 exhibited low CO2 capture capacity. The weight of the sample decreased 0.11% (0.0055 mg).

With correction, the weight gain became 0.57% which was higher than method 1 result. Hence, it was safe to assume method 2 resulted in more oxidized functional groups, and thus increased CO2 capacity.

Figure 15. TGA result for 30W GO F1

With correction, 30W GO F2 had a 1.38% weight increase which was around twice as much as 30W GO F1 (0.57%). Clearly, longer method time resulted in more aminated functional groups (Figure 16).

Figure 16. TGA result for 30W GO F2

With 30W GO F3, the corrected weight percent gain was 1.35%. It was a significant increase compared to the 30W GO F1 (0.57%) but a slight decrease compared to 30W GO F2

(1.38%). The weight change can due to human error since it was relatively insignificant (2.4% difference). Obviously, both the 72 hour method time and the 144 hour method time have an advantage over the 24 hour method time. Hence, for the functionalization process with second purification method (4W graphite oxide), the 24 hour method time was neglected and only the 72 hour and the 144 hour method times were tested.

Figure 17 TGA result for 30W GO F3

III.2.2 Method 2 with Purification 2 (four washes)

As shown in Figure 18, the FTIR spectrum exhibited the same peaks as method 1 with very strong peaks at 1614 cm-1, 1351 cm-1, and 1053 cm-1. The 1614 cm-1 peak could be a N-H bend for primary amines. The peak at 1053 cm-1 could be a C-N stretch for aliphatic amines

(nitrogen not attached to benzene ring, amine functionalized epoxy and carboxylic acid group).

However, the primary peak (N-H stretch for primary amine, see Figure 19) overlapped with O-H stretch (strong, broad); it was difficult to determine the existence of such a peak. The C=O stretch for carboxylic acid is around 1650 cm-1; it might overlap with the N-H bend, and the C-O stretch for carboxylic acids (strong) could overlap with the C-N stretch (9).

Figure 18 FTIR spectrum for method 2 purification 2, 4W GO F1 has method time 72 hours and 4W GOF2 has method time 144 hours.

The CO2 capture capacity of this sample exhibited a decrease compared to 30W GO. As shown below (Figure 19), this sample captured around 1.05% weight of CO2, which was 23.7% lower compared to the 30W GO F2 (same method time).

With a much longer method time (144 hours), there was an increase in CO2 capture capacity with a corrected weight increase at 1.2% (Figure 20) compared to 4W GO F1 (72 hours).

This meant more nitrogen was attached to the functional groups during the extra period of time.

The sample with 144 hours method time illustrated only a 14% increase in terms of CO2 capture capacity.

Figure 19

Figure 20

CHAPTER IV

CONCLUSION AND FUTURE WORK

Method 2 with 30 washes had an advantage in terms of CO2 capture capacity compared to method 1 and method 2 with 4 washes. However, the difference was only 10.3% (compared the 30W GO F3 to 4W GO F2). To achieve a stable pH, it took three days to wash graphite oxide.

Four washes were much easier to achieve as it took much less time. This could also indicate that the impurities in graphite oxide did not have a large effect on the attachment of amine onto the functional groups, or four washes was sufficient enough to remove most of the impurities. From

Figure 23, compared to the 4W GO F1, there was an increase in CO2 capture with the 4W GO F2

(144 hours). This could mean the maximum potential of amine attachment was not achieved and perhaps a longer method time should be tested.

As shown in TGA plots, all samples exhibited an increasing exponential decay curve

(f(t) = Ae−B/t + C, where A, B, and C are constants) which means the sorbent reached 90% of its maximum capacity within the first three time constants. Note the 30WGO had larger maximum CO2 capture capacity than the 4WGO, but the reaction rate was much slower. 30W

GO F2 required 30 minutes and 30W GO F3 required 60 minutes to reach stabilized weight. The

4W GO F1 only used 8 minutes and the 4W GO F2 only used 9 minutes to reach its maximum

CO2 capacity. Hence, there was a tradeoff between CO2 capture rate and CO2 capture capacity between the two samples.

To test the feasibility of the functionalized GO in the industry, it is recommended to gather more precise data by using 15% CO2 instead of pure CO2. Pure CO2 has a much higher driving force which is unrealistic from the industry stand point. Moreover, Gas chromatography–mass spectrometry (GC/MS) should be used to plot the breakthrough curve which can show time required for CO2 to pass through a sorbent packed bed (7). A breakthrough curve can give further evidence of CO2 capture as well as the sorbent weight required to capture a given CO2 emission rate.

Figure 21. Example of Breakthrough curves (15) Additionally, the effect of ultrasound should be studied. Graphite oxide can be broken into graphene oxide with either ultrasound or physical force. The optimum ultrasound time, power level, and frequency should be determined not only to successfully break down graphite oxide without damaging functional groups but also minimize the energy output.

Compared to method 1, the functionalized GO in method 2 had much more intensive peaks at 1614 cm-1 and 1351 cm-1 but a less intensive peak at 1058 cm-1. Therefore, the FTIR spectrum did not explain why method 2 captured more CO2. It would be much more accurate if elemental analyses were performed on all the samples to quantify the ratio of N, C, and O presented in the functionalized GO sample.

Based on weight gain, the C:O molar ratio was around 2.1:1 (see Figure 23 and Eq1) which agreed with the literature (14). With the assumption that 60% of all functional groups were hydroxyl groups, there was additional 20 mole percent gain of oxygen in epoxy and carboxylic acid groups. In theory, both functional groups should be aminated to capture CO2.

However, this was not the case. From TGA results, it was clear that the functionalization ratio reached equilibrium since there was only a small percent gain (14%) after additional 72 hours in an amine rich solution. This can be due a reverse reaction—amine detached from the functional groups (during stirring). The most convenient variables to manipulate in order to improve convergence rate are concentration of amine and the temperature of the solution. A higher concentration of amine can create a greater driving force and could force amine to attach to the functional groups. Higher temperature usually increases the reaction rate; however, with higher temperature, there might be a greater reverse reaction. Therefore, a few experiments with temperature variation should be plotted and the reverse reaction rate calculated by using nonlinear regression. It is also recommended to plot the method time verses CO2 capture capacity. Based on collected date, the curve might be an increasing exceptional decay curve

(f(t) = Ae−B/t + C), but more data should be collected in order to support this theory.

This experiment has focused on absorption, not adsorption. Although the surface area of graphene oxide was relatively high at 63 m2/g (Figure 27), it is much smaller compared to GAC

(granular activated carbon) which can be around 500 m2/g (3). Graphene oxide is also in single to few layers, which means it is not structured to trap CO2 in pores. We could combine GAC and

GO to take advantage of adsorption and absorption possibly yielding a greater CO2 capture capacity.

REFERENCE

1. Global Climate Change: Evidence. (2014). Retrieved March 6, 2015, from http://climate.nasa.gov/evidence/

2. Carbon Dioxide Emissions (2012). Retrieved March 6, 2015, from http://www.epa.gov/climatechange/ghgemissions/gases/co2.html

3. Yang, H., Xu, Z., Fan, M., Gupta, R., Slimane, R. B., Bland, A. E., & Wright, I. (2008). Progress in carbon dioxide separation and capture: A review. Journal Of Environmental Sciences, 2014-27. doi:10.1016/S1001-0742(08)60002-9

4. Marcano, D., Kosynkin, D., Berlin, J., Sinitskii, A., Sun, Z., Slesarev, A., ... Tour, J. (2010). Improved Synthesis of Graphene Oxide. ACS Nano, 4(8), 4806-4814.

5. Stankovich, S., Dikin, D., Dommett, G., Kohlhaas, K., Zimney, E., Stach, E., ... Ruoff, R. (2006). Graphene-based composite materials. Nature, 422(20), 282-286.

6. Dreyer, D., Park, S., Bielawski, C., & Ruoff, R. (2009). The Chemistry Of Graphene Oxide. Chemical Society Reviews, 39, 228-240.

7. Zhao, Y., Ding, H., & Zhong, Q. (2012). Preparation and characterization of animated graphite oxide for CO2 capture. Applied Surface Science, 258, 4301-4307.

8. Whitby, R., Korobeinyk, A., & Glevatska, K. (2011). Morphological changes and covalent reactivity assessment of single-layer graphene oxides under carboxylic group- targeted chemistry. Carbon, 49(2), 722-725. doi:10.1016/j.carbon.2010.09.049

9. Hanson, John. "IR-frequencies." IR-frequencies. University of Puget Sound, 24 June 2002. Web. 01 Apr. 2015. http://www2.ups.edu/faculty/hanson/Spectroscopy/IR/IRfrequencies.html

10. Zou, Zheng-Guang, Hui-Jiang Yu, Fei Long, and Yan-Huang Fan. "Preparation of Graphene Oxide by Ultrasound-Assisted Hummers Method." Chinese Journal Of Inorganic Chemistry 27.09 (2011)

11. Chen, W., Mattern, D., Okinedo, E., Senter, J., Mattei, A., & Redwine, C. (2014). Photochemical and acoustic interactions of biochar with COand HO: Applications in power generation and COcapture. AIChE Journal, 60(3), 1054-1065. Retrieved April 28, 2015, from wileyonlinelibrary.com

12. Park S, An J, Piner RD, Jung I, Yang D, Velamakanni A, Nguyen ST, Ruoff RS. Aqueous suspension and characterization of chemically modified graphene sheets. Chem. Mater. 2008; 20: 6592-6594.

13. Stankovich, S., Dikin, D., Piner, R., Kohlhaas, K., Kleinhammes, A., Jia, Y., . . . Ruoff, R. (2007). Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite oxide. Carbon, 45(7), 1558-1565.

14. Pei, S., & Cheng, H. (2011). The reduction of graphene oxide. Carbon, 50(9), 3210-3228.

15. Corrêa, R. A., Calçada, L. A., & Peçanha, R. P.. (2007). Development of a fluidized bed system for adsorption of phenol from aqueous solutions with commercial macroporous resins. Brazilian Journal of Chemical Engineering, 24(1), 15-28. Retrieved April 28, 2015, from http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104- 66322007000100002&lng=en&tlng=en. 10.1590/S0104-66322007000100002

APPENDIX

APPENDIX A: ADDITIONAL INFORMATION FOR METHOD 1

pH of Graphite Oxide After Each Wash # of Washes 1 2 3 4 5 6 7 8 9 pH of GO 0.13 0.74 1.56 1.91 2.22 2.51 2.55 3.01 3.15 # of Washes 10 11 12 13 14 15 16 17 18 pH of GO 3.27 3.3 3.4 3.41 3.5 3.45 3.52 3.5 3.51 Table 1

Figure 22. FTIR spectrum of pure EDA

APPENDIX B: ADDITIONAL INFORMATION FOR METHOD 2

Weight Change 2.3

2.2

2.1

1.9

1.8

1.7

Graphite Oxide Weight, g Weight, Oxide Graphite 1.6

1.5 0 2 4 6 8 10 12 Time, days

Figure 23

Approximated C:O ratio: Assume hydrogen is negligible and all weight gain is oxygen.

massGraphite /MC C: O = (Eq 1) (massGO − massGraphite)/MO

mol 1g 12 g molC C: O = g = 2.083 (1.64g − 1 g)/(16 ) molO mol

APPENDIX C: TGA CORRECTION

Figure 24, Blank run with empty pan

Mass loss:

m = .3747 ∗ 0.0900 mg = 0.03372 mg

Figure 25. Blank run with weight Mass loss:

m = 0.001642 ∗ 20.9789 mg = 0.03445 mg

Percent weight difference:

0.03445mg − 0.03372 mg ∗ 100% = 2.143% 0.03445mg

Sample Calculation:

4.953 mg ∗ .0011 + 0.0339mg ∗ 100% = 0.57% (Eq 2) 4.953 mg

Weight Gain with Correction Sample Sample Weight, mg Weight gain % Corrected % Gain 30W GO F1 4.953 -0.11 0.5704 30W GO F2 4.726 0.6667 1.380 30W GO F3 11.382 1.052 1.348 4W GO F1 35.352 0.9565 1.052 4W GO F2 27.443 1.073 1.196 Table 2

1.6

1.4

1.2

0.8 GO 30W 0.6 GO 4W 0.4

Corrected Weight PercentGain 0.2

0 0 20 40 60 80 100 120 140 160 Method Time, Days

Figure 26

APPENDIX D: SURFACE AREA

1.2 Muli-Point BET Plot 1

0.8

1)] - 0.6

0.4 1/[W((Po/P) 0.2

0 -10 0 10 20 30 40 50 60 70 -0.2 Relative Pressure, P/Po

BET summary Slope = 60.082 Figure 27, surface area calculated by using Brunauer– Intercept = -5.037e+00 Emmett–Teller (BET) theory with Nova 2000. Correlation coefficient, r = 0.966674 C constant= -10.929 Surface Area = 63.267 m2/g

APPENDIX E: ADDITIONAL FTIR INFORMATION

Figure 28

Figure 29

VITA

RENEE RAN WEI ZELESZKI EDUCATION: M.S., Engineering Science, The University of Mississippi, to be awarded August 2015. Research Area: Amine Functionalized Graphene Oxide for Carbon Dioxide Capture

B.S., Chemical Engineering, University of California, Davis, Aug 2009.

CERTIFICATE: Engineer-in-Training, License # EIT 135913, CA, July 2009 WORK EXPERIENCES: Graduate Assistant /Researcher, 2013-2015 The University of Mississippi

Scientific Aide, 2010-2011 California Dept of Food and Agriculture/CASS

ASSOCIATION: AICHE, 2007-2009